1. Field of the Invention
This invention relates to a digital signal decoding apparatus adapted for decoding a transmitted or reproduced signal (data) obtained after a signal coded by an efficient coding apparatus for efficiently coding input digital data by the so-called block floating processing is transmitted or recorded.
2. Description of the Prior Art
There has been conventionally known an efficient coding technology to orthogonally transform an audio signal, etc. to divide its orthogonally transformed output into blocks every plural words to carry out floating processing every respective blocks to implement quantization thereto to record on a medium, or transmit thereto, floating information and quantization information along with a quantized output. Here, the above-mentioned block floating technique is basically directed to a technique to multiply respective words within the block by a value common thereto so that they have larger values, thus to improve the accuracy at the time of quantization. For an actual example, there is a block floating technique to search for the maximum one of absolute values (i.e., maximum absolute value) of respective words within the block to implement floating processing to all words within the block by using a floating coefficient common thereto, so that the maximum absolute value is not saturated. For a simpler block floating technique, there is also a block floating technique with 6 dB being as a unit utilizing bit shift.
However, in conventional orthogonal transform processing, without use of the block floating technique, a sufficient accuracy is ensured even at the time of any input, and an operation word length is made long to such a sufficient degree that the word length accuracy of a signal inputted to an orthogonal transform circuit is not damaged by the orthogonal transform processing. Further, an approach is employed to allow the block size for the orthogonal transform processing to be variable depending upon the property in point of time of a signal, thus to improve the analysis accuracy. There are instances where, as an index for judgment therefor, root mean square values of differences between adjacent samples of a signal are used.
Meanwhile, in such orthogonal transform processing, in order to perform an operation while maintaining the accuracy of an input, the operation word length becomes longer. For this reason, the scale of the hardware becomes large, resulting in great difficulty from an economical point of view. Further, in the case where the block size for the orthogonal transform processing is caused to be variable, newly determining a judgment index therefor only for the purpose results in an increase in the number of operation steps.
Further, in order to search for the above-described maximum absolute value in the above-mentioned block floating processing, such a procedure is required to judge whether or not the absolute value of the present (current) word is larger than the maximum absolute value of past words with respect to all words within one block. As a result, the number of steps in the processing program becomes large and it, takes much time therefor.
In view of the above, the applicant of this invention has already proposed, in the specification of the Japanese Patent Application No. 235613/1991 and the drawings attached therewith, an efficient coding technique to orthogonally transform an audio signal, etc. to divide the orthogonally transformed output, into blocks every plural words to carry out floating processing every respective blocks to implement quantization thereto to record floating information and quantization information along with a quantized output onto a medium, or transmit such information thereto wherein an approach is employed to carry out a first block floating processing before the orthhogonal transform processing, and to carry out a second block floating processing after the orthogonal transform processing to thereby prevent degradation in the operation accuracy during the execution of the orthogonal transform operation. This technique which has been already proposed in the specification and the drawings mentioned above is directed to the technique to multiply respective words within a unit to carry out the orthogonal transform processing (orthogonal transform block) by a coefficient common thereto, thus to improve the accuracy at the time of the orthogonal transform processing. As a simpler method, a block floating with 6 dB being as a unit utilizing bit shift is frequently used.
Meanwhile, in the inverse orthogonal transform processing for decoding corresponding to the orthogonal transform processing in the above-mentioned efficient coding, when an attempt is made to implement inverse orthogonal transform processing to words which have been subjected to floating before inputting to the orthogonal transform circuit while maintaining its floating state, there results a high possibility that an overflow may take place in the process of the operation.
In order to prevent such overflow, however, when small floating coefficients are used to carry out floating, there results degrated operation accuracy.